INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

HEARTS EXPOSED to a brief period of ischemia followed by reperfusion become more tolerant to subsequent ischemic episodes. This phenomenon is called ischemic preconditioning (31). Preischemic stimulation of receptors by various agonists such as adenosine, acetylcholine, catecholamines, bradykinin, angiotensin II, endothelin-1, and opioids are also reported to protect the myocardium exposed to ischemia-reperfusion (I/R) (8). Although activation of protein kinase C (PKC) has been implicated as an important downstream pathway through which multiple triggers may act (25, 47), it is not clear what the end effector is that protects the myocardium.

PKC has been reported to activate Na⁺/H⁺ exchanger and vacuolar proton ATPase (32, 34, 46). If Na⁺/H⁺ exchanger and vacuolar proton ATPase are activated during ischemia, intracellular acidosis during ischemia would be attenuated, which might elicit cardioprotective effects (13). On the other hand, if these pH-regulatory mechanisms are activated during reperfusion, the intracellular pH (pH_i) would return more rapidly toward the physiological level. This rapid pH_i recovery, however, might not be beneficial to reperfused hearts, according to studies reporting that acidosis during early reperfusion limited myocardial injury during I/R (23, 24, 33).

Because the cardioprotective effects by ischemic preconditioning or PKC activation were reported not only in in vivo models and isolated heart preparations but also in isolated myocyte models (4, 35), myocytes themselves seem to possess the cardioprotective mechanism. Although depolarization of sarcolemmal membranes affects intracellular ion mobilizations, there has been no report concerning the effects of PKC preactivation on cell motion and ionic alterations during I/R in isolated myocytes that were electrically stimulated throughout the course of the experiment.

The purpose of this study was to examine the effects of preactivation of PKC on cell motion and ionic alterations during simulated I/R in isolated, paced rat ventricular myocytes. We measured the cell motion, intracellular Ca²⁺ concentration ([Ca²⁺]_i), and pH_i using the fluorescence indicators indo 1 and seminaphthorhodafluor-1 (SNARF 1) in collagenase-dissociated paced rat ventricular myocytes, because [Ca²⁺]_i and pH_i have been shown to play important roles in cell damage (6, 37). The results of the present study indicate that preactivation of PKC attenuates the I/R injury in isolated, paced ventricular myocytes and is accompanied by a delayed overshoot in pH_i during reperfusion through a PKC-dependent mechanism(s).

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Dissociation of left ventricular myocytes. Left ventricular myocytes were dissociated from male Wistar rats (160-180 g body wt) as described previously (18-21). Rats were anesthetized with an intraperitoneal injection of pentobarbital sodium (50 mg/kg body wt). The heart was rapidly excised and attached to an aortic cannula. Continuous retrograde coronary perfusion was then initiated at a perfusion pressure of 70 cmH₂O. The heart was first perfused with nominally Ca²⁺-free modified Krebs-Henseleit buffer of the following composition (in mM): 123 NaCl, 5.4 KCl, 1.2 MgSO₄, 1.2 NaH₂PO₄, 20 NaHCO₃, and 11 glucose. This medium was not recirculated and was continuously gassed with 95% O₂-5% CO₂ (pH 7.4, 36-37°C). After 3 min of the initial perfusion, the heart was perfused with recirculating Krebs-Henseleit buffer supplemented with 0.08 mg/ml collagenase P (Boehringer Mannheim, Mannheim, Germany), 0.04 mg/ml protease (type XIV; Sigma Chemical, St. Louis, MO), and 1 mg/ml BSA (Sigma Chemical) for 40-50 min. The heart was then detached from the cannula. The left ventricle was cut into small pieces, and the dispersion of the myocytes was performed by gentle agitation of the ventricular tissue through a serologic pipette in Krebs-Henseleit buffer containing 100 µM CaCl₂ and 1 mg/ml BSA. The resulting suspension was then gently forced through a 450-µm nylon screen filtration cloth into a 50-ml plastic tube and rinsed twice. The myocytes were then resuspended in HEPES-buffered solution of the following composition (in mM): 137 NaCl, 5.4 KCl, 1.2 MgSO₄, 1.2 NaH₂PO₄, 20 HEPES (free acid), 1.2 CaCl₂, 20 glucose, and 5 mg/ml BSA. The myocytes were stored at 36°C for 1 h.

Simultaneous measurement of [Ca²⁺]_i and cell contraction. [Ca²⁺]_i was measured with the Ca²⁺-sensitive fluorescence indicator indo 1-AM (Dojindo Laboratories, Kumamoto, Japan) (14) as described previously (18-21) using a modification of the method of duBell et al. (10) and Ikenouchi et al. (17). Myocytes were loaded with 2 µM indo 1-AM in HEPES-buffered solution at room temperature for 20 min. The coverslip was rinsed with indo 1-AM-free buffer solution and placed in a flow-through heated (36-37°C) cell superfusion chamber on the stage of an inverted microscope (Nikon, Tokyo, Japan). The instrumentation for fluorescence measurement was prepared as described by Peeters et al. (36). The excitation source was a high-pressure mercury arc lamp, which provides an intense emission peak at 360 nm. Further selection of this excitation was made with narrow-bandwidth interference filters. The myocyte was illuminated via epifluorescent optics using a Fluor ×40 objective lens (Nikon). The fluorescence light was collected by the objective lens and transmitted to a spectrofluorometer (CAM-220; Nikon) for simultaneous measurement of both 400- and 500-nm wavelengths using two separate photomultiplier tubes. The spectrofluorometer provided analog signals representing the fluorescence intensity at both wavelengths. After background autofluorescence obtained from an unloaded myocyte was subtracted at the end of each experiment, the ratio of emitted fluorescence (F₄₀₀/F₅₀₀) was calculated. An adjustable rectangular window was used to restrict the optical image to only one myocyte of interest in each experiment to minimize background fluorescence from other cells and debris. The image of the beating myocyte was obtained by illumination via a 50-W standard microscope light source passed through a 645-nm band-pass filter. This wavelength was long enough not to interfere with the fluorescence detection at 400 and 500 nm. The motion of the myocyte was monitored using a charge-coupled device camera (TM-640; PULNiX, Sunnyvale, CA) and a custom-modified video detector system (Crescent Electronics, Sandy, UT) (43). The analog output of the cell motion signal was monitored and recorded continuously with the analog signal of the [Ca²⁺]_i-sensitive fluorescence ratio (F₄₀₀/F₅₀₀). Two platinum electrodes placed in the bathing fluid were connected to a stimulator (SEN-3201; Nihon Kohden, Tokyo, Japan) and used to stimulate the myocyte at 1.25 Hz with 3-ms pulses.

Measurement of pH_i. The pH_i was measured with the pH-sensitive indicator SNARF 1-AM (Molecular Probes, Eugene, OR) as described recently (18, 19). First, 50 µg of SNARF 1-AM was added to 50 µl of DMSO, which was mixed with 450 µl of HEPES-buffered solution containing 5 mg/ml BSA. Loading of SNARF 1 AM was done by exposing the myocytes on coverslips to a final concentration of 4 µM SNARF 1-AM for 20 min at room temperature. The coverslip was then rinsed with SNARF 1-free solution and placed on a flow-through cell superfusion chamber as described in Simultaneous measurement of [Ca²⁺]_i and cell contraction. Excitation was performed at 540 nm, and fluorescence emission was collected simultaneously at 580 and 640 nm using the same optics system described above except for the substitution of different dichroic mirrors and interference filters. At the end of each experiment, the myocytes were superfused with HEPES-buffered solution containing 500 nM thapsigargin until the beating disappeared, followed by exposure to Ca²⁺-free HEPES-buffered solution containing 30 mM 2,3-butanedione monoxime (BDM), 2 mM EGTA, and 1 µM thapsigargin for 10 min. The emission ratio from each cell was then calibrated in situ by exposing the cells to solutions of varying pH. Each solution contained (in mM) 140 K⁺ (adjusted to keep extracellular K⁺ concentration constant), 1.0 MgCl₂, 4.0 HEPES, 40 BDM, 10 EGTA, and 11 glucose as well as 14 µM nigericin (free acid) (Molecular Probes) and was titrated to various pH values (6.4, 6.7, 7.0, 7.3, 7.6, 7.8) using 1.0 N KOH. Because changes in pH are not linearly related to the ratio of fluorescence emission (42), pH_i was calculated by the equation pH_i = (ax + c)/(1 + bx), where x is the measured emission ratio and a, b, and c are constant parameters (18, 19, 30, 42). Because the fluorescence from unloaded cells was <1% of the fluorescence signal from SNARF 1-loaded cells and was almost the same level as that from a cell-free area, the background fluorescence at each wavelength was measured as the fluorescence from a clean, cell-free area next to the cell under study. This fluorescence was subtracted from the loaded-cell signal. There was no change in the background fluorescence at each wavelength during simulated I/R.

Effects of 1,2-dioctanoyl-sn-glycerol on cell contraction and [Ca²⁺]_i. All experiments in the present study were performed with myocytes that were continuously paced at 1.25 Hz throughout the course of the experiment. The temperature of the myocytes was maintained at 36-37°C. In this protocol, myocytes from 14 hearts were studied. Two to four experiments were performed in sequence from separate coverslips of myocytes isolated from one heart. The myocytes that were loaded with indo 1 were superfused with HEPES-buffered solution (control buffer) of the following composition (in mM): 130 NaCl, 4.0 KCl, 1.0 MgSO₄, 1.2 CaCl₂, 10 HEPES (Na salt), and 11 glucose, with a final pH of 7.40. Probenesid (0.5 mM), a blocker of organic anion transport, was added to all solutions because it has been shown to inhibit the secretion of both indo 1 and fura 2 free acids from loaded cells (3, 9). A membrane-permeable diacylglycerol, 1,2-dioctanoyl-sn-glycerol (DOG; Sigma Chemical), was used as an activator of PKC (7). DOG (5 mg) was dissolved with 0.1 ml of DMSO and then aliquoted into 10-µl samples, which were stored at 70°C. In the control group without pretreatment with DOG, the same amount of DMSO was added as a substitute for DOG. The final concentration of DMSO was <0.01% . The experimental protocol is shown in Fig. 1. The myocytes were first paced at 0.5 Hz, and then the pacing rate was gradually increased to 1.25 Hz. After the myocyte twitch became stable, the paced myocytes were superfused with the control buffer containing 10 µM DOG (n = 19, DOG group) or DMSO (n = 23, control group) for 5 min, followed by superfusion with the control buffer for 5 min. They were then exposed to the simulated ischemia buffer for 3 min and were reperfused with the control buffer for 10 min. The composition of the simulated ischemia buffer was as follows (in mM): 110 NaCl, 12 KCl, 1.0 MgSO₄, 1.2 CaCl₂, 10 HEPES (Na salt), 10 2-deoxyglucose, 1.5 NaCN, and 20 sodium lactate, with a final pH of 6.50. This buffer was chosen to simulate the extracellular milieu of reversible myocardial ischemia (11, 33). This duration of simulated ischemia was selected because, when the duration was more than 5 min, almost none of the cells could keep the rod shape after reperfusion. The analog cell motion signals and the F₄₀₀/F₅₀₀ analog signals were recorded simultaneously.

View larger version (15K): [in this window] [in a new window]   Fig. 1.   Representative tracings of cell motion in myocytes pretreated without (A) or with (B) 10 µM 1,2-dioctanoyl-sn-glycerol (DOG). Myocyte shortening is displayed as an upward deflection of the cell motion trace. In both myocytes, twitch amplitude was reduced during simulated ischemia. Immediately after the onset of reperfusion, diastolic cell length was more shortened in the myocyte without DOG pretreatment than in the myocyte pretreated with DOG.

Effects of DOG on cell contraction and pH_i. In this protocol, myocytes from 15 hearts were studied. Two to four experiments were performed in sequence from separate coverslips of myocytes isolated from one heart. After the myocyte twitch became stable, the paced myocytes loaded with SNARF 1 were superfused with the control buffer containing 10 µM DOG (n = 17, DOG group) or DMSO (n = 17, control group) for 5 min, followed by superfusion with the control buffer for 5 min. They were then exposed to the simulated ischemia buffer for 3 min and were reperfused with the control buffer for 10 min. The analog cell motion signals and the F₆₄₀/F₅₈₀ analog signals were recorded. Because the dichroic mirrors used in our system do not allow the measurement of both cell motion and fluorescence signals from SNARF 1-loaded myocytes simultaneously, fluorescence signals were recorded for 8-10 s at baseline, at intervals of 5 min before ischemia, at intervals of 1 min during I/R, and continuously during the first minute after the onset of reperfusion. Cell motion signals were obtained when the fluorescence signals were not being measured.

Effects of PKC inhibition. To assess the contribution of PKC activation to the effects induced by DOG, another experiment was performed using a highly selective inhibitor of PKC, chelerythrine chloride (Research Biochemicals International, Natick, MA) (16). In this protocol, myocytes from 11 hearts were studied. Two to four experiments were performed in sequence from separate coverslips of myocytes isolated from one heart (see experimental protocol shown in Fig. 8). After the myocyte twitch became stable, the myocytes that were loaded with SNARF 1 were superfused with the control buffer alone (n = 10, control group) or control buffer containing 2 µM chelerythrine (n = 10, Che-DOG group; n = 8, Che group) for 5 min. The myocytes were then superfused with the control buffer containing 10 µM DOG (Che-DOG group) or DMSO (control group, Che group) for 5 min, followed by superfusion with the control buffer for 5 min. They were then exposed to the simulated ischemia buffer for 3 min and were reperfused with the control buffer for 10 min. In the Che-DOG group and the Che group, the myocytes were exposed to chelerythrine for 15 min before simulated ischemia. The analog cell motion signals and the F₆₄₀/F₅₈₀ analog signals were recorded.

Effects of BDM on the occurrence of cell contracture. To evaluate whether the reperfusion-induced contracture could be observed in the presence of BDM, which uncouples the contractile activity from Ca²⁺ transients (15, 38), an additional experiment was performed using myocytes without dye loading (BDM protocol). In this protocol, myocytes from five hearts were studied. Two to four experiments were performed in sequence from separate coverslips of myocytes isolated from one heart. After the myocyte twitch became stable, the paced myocytes were superfused with the control buffer containing 10 µM DOG (n = 8, DOG group) or DMSO (n = 8, control group) for 5 min, followed by superfusion with the control buffer containing 10 mM BDM for 5 min. They were then exposed to the simulated ischemia buffer containing BDM for 3 min and were reperfused with the control buffer containing BDM for 10 min. The analog cell motion signals were recorded.

Statistical analysis. Two-way ANOVA with repeated measures was used to compare the values measured during simulated I/R between the control and DOG groups or among the three groups in the protocol using chelerythrine. Unpaired Student's t-test was used for comparisons of the baseline data between the control and DOG groups. Comparisons of baseline data among the three groups in the chelerythrine protocol were performed using ANOVA. Spearman's rank correlation was used to evaluate the correlation between the degree of reperfusion-induced contracture and the time of pH recovery during reperfusion. P < 0.05 was considered significant. Results are expressed as means ± SE.

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

Effects of DOG on cell contraction and [Ca²⁺]_i. Representative recordings of cell motion in myocytes pretreated with or without 10 µM DOG are shown in Fig. 1. In both myocytes, the twitch amplitude was reduced during simulated ischemia. Immediately after the onset of reperfusion, the twitch amplitude rapidly increased and the diastolic cell length was shortened. The diastolic cell length was more shortened in the myocyte without DOG pretreatment than in the myocyte pretreated with DOG. There were no differences in diastolic cell length, twitch amplitude, or [Ca²⁺]_i at baseline, just before the exposure to ischemia, or during ischemia between the DOG group (n = 19) and the control group (n = 23) (Figs. 2-4). The decrease () in diastolic cell length during reperfusion was significantly attenuated in the DOG group compared with the control group (3.3 ± 1.2 vs. 14.4 ± 4.5% of baseline diastolic cell length, P < 0.02) (Fig. 2). However, there were no differences in twitch amplitude, end-diastolic [Ca²⁺]_i, or peak-systolic [Ca²⁺]_i during reperfusion between the two groups (Figs. 3 and 4). These data suggest that the pretreatment with DOG attenuates the myocardial damage during simulated I/R in isolated, paced myocytes and that its mechanism cannot be explained by the attenuation of the cytosolic Ca²⁺ overload.

View larger version (25K): [in this window] [in a new window]   Fig. 2.   Changes in end-diastolic cell length during ischemia-reperfusion (I/R) in control (, n = 23) and DOG-pretreated (, n = 19) myocytes loaded with indo 1. Values are normalized relative to baseline for each myocyte and are means ± SE. In both groups, diastolic cell length was increased to some degree after superfusion with simulated ischemia buffer and was shortened immediately after the onset of reperfusion. The magnitude of contracture was significantly attenuated in the DOG group compared with the control group (P < 0.02).

View larger version (24K): [in this window] [in a new window]   Fig. 3.   Changes in twitch amplitude during I/R in control (, n = 23) and DOG-pretreated (, n = 19) myocytes loaded with indo 1. Values are normalized relative to baseline diastolic cell length for each myocyte and are means ± SE. In both groups, twitch amplitude decreased after superfusion with simulated ischemia buffer. Immediately after the onset of reperfusion, twitch amplitude transiently increased and then gradually returned toward the control value. There were no significant differences in twitch amplitude during I/R between the 2 groups.

View larger version (31K): [in this window] [in a new window]   Fig. 4.   Changes in end-diastolic (filled symbols) and peak-systolic (open symbols) intracellular Ca2+ level during I/R in control (circles, n = 23) and DOG-pretreated (triangles, n = 19) myocytes loaded with indo 1. Vertical axis represents intracellular Ca2+ concentration ([Ca2+]i) as recorded with indo 1. Values are means ± SE. In both groups, Ca2+ transient amplitude decreased after superfusion with simulated ischemia buffer. Immediately after the onset of reperfusion, Ca2+ transient amplitude increased and then gradually decreased. There was no difference in either end-diastolic or peak-systolic [Ca2+]i during I/R between the 2 groups.

Effects of DOG on cell contraction and pH_i. The pH_i is known to modulate the myofilament kinetics and responsiveness to Ca²⁺. Therefore, the changes in pH_i during simulated I/R were studied. There were no differences in diastolic cell length (Fig. 5) or twitch amplitude (data not shown) at baseline, just before the exposure to ischemia, or during ischemia between the DOG group (n = 17) and the control group (n = 17). The decrease in diastolic cell length during reperfusion was significantly attenuated in the DOG group compared with the control group (29.2 ± 4.5 vs. 50.4 ± 6.5% of baseline diastolic cell length, P < 0.001) (Fig. 5). Although there were no statistical differences in pH_i before the exposure to ischemia or during ischemia between the two groups (n = 17 for each), the pH_i recovery and the subsequent overshoot during reperfusion in the DOG group were significantly delayed compared with those in the control group (P < 0.005) (Fig. 6). There was no difference in twitch amplitude (data not shown) or in the peak pH_i (7.57 ± 0.05 vs. 7.64 ± 0.06, respectively, P = 0.38) during reperfusion between the DOG-pretreated and untreated myocytes. The correlation between the diastolic cell length (%baseline) 10 min after the reperfusion and the time to pH_i 7.30 from the onset of reperfusion was statistically significant (n = 34, r = 0.44, P < 0.0005) (Fig. 7). These data suggest that PKC preactivation attenuates I/R injury in isolated, paced myocytes and that this protective effect may be related to the slower pH_i recovery and subsequent overshoot during reperfusion.

View larger version (24K): [in this window] [in a new window]   Fig. 5.   Changes in end-diastolic cell length during I/R in control (, n = 17) and DOG-pretreated (, n = 17) myocytes loaded with seminaphthorhodofluor 1 (SNARF 1). Values are normalized relative to baseline for each myocyte and are means ± SE. In both groups, diastolic cell length increased to some degree after superfusion with simulated ischemia buffer and was shortened immediately after the onset of reperfusion. The magnitude of contracture was significantly attenuated in the DOG group compared with the control group (P < 0.001).

View larger version (39K): [in this window] [in a new window]   Fig. 6.   A: changes in intracellular pH (pHi) during I/R in control (, n = 17) and DOG-pretreated (, n = 17) myocytes loaded with SNARF 1. B: first 10 min of reperfusion from A. Values are means ± SE. In both groups, pHi decreased after superfusion with simulated ischemia buffer. After the onset of reperfusion, pHi increased and then gradually decreased in both groups. Although there was no difference in pHi during ischemia, pHi recovery and subsequent overshoot during reperfusion were significantly delayed in the DOG group compared with the control group (P < 0.005).

View larger version (13K): [in this window] [in a new window]   Fig. 7.   Relationship between diastolic cell length (%baseline) 10 min after the onset of reperfusion and time to pHi 7.30 from the onset of reperfusion (n = 34). There was a positive correlation between the 2 parameters (r = 0.44, P < 0.0005).

Effects of PKC inhibition. To investigate the contribution of PKC activation to the cardioprotective effect by DOG, an additional experiment using chelerythrine was performed. There were no differences in diastolic cell length (Fig. 8), twitch amplitude (data not shown), or pH_i (Fig. 9) at baseline among the three groups (n = 10 for control group; n = 10 for Che-DOG group, n = 8 for Che group). There were no differences in diastolic cell length during reperfusion among the three groups (Fig. 8). Moreover, there were no differences in the pH_i during I/R among the three groups (Fig. 9). Both the attenuation of the reperfusion-induced contracture and the delayed pH_i recovery and subsequent overshoot elicited by DOG pretreatment were completely abolished by the PKC inhibitor chelerythrine, suggesting that the favorable effects of DOG were PKC dependent. In the groups treated with chelerythrine (Che-DOG group and Che group), pH_i increased significantly after the superfusion with chelerythrine (P < 0.005 vs. control group) (Fig. 9A). These changes suggest the existence of a PKC-dependent mechanism that lowers the pH_i.

View larger version (23K): [in this window] [in a new window]   Fig. 8.   Changes in end-diastolic cell length during I/R in control myocytes (, n = 10), myocytes pretreated with both DOG and chelerythrine (, n = 10), and myocytes pretreated with chelerythrine (, n = 8) that were loaded with SNARF 1. Values are normalized relative to baseline for each myocyte and are means ± SE. There were no significant differences in diastolic cell length during I/R among the 3 groups. The protective effect of DOG on diastolic cell length was completely abolished by pretreatment with chelerythrine.

View larger version (44K): [in this window] [in a new window]   Fig. 9.   A: changes in pHi during I/R in control myocytes (, n = 10), myocytes pretreated with both DOG and chelerythrine (, n = 10), and myocytes pretreated with chelerythrine (, n = 8) that were loaded with SNARF 1. B: first 10 min of reperfusion from A. Values are means ± SE. In both groups treated with chelerythrine, pHi increased significantly after superfusion with chelerythrine. There were no significant differences in pHi during I/R among the 3 groups. Slower pHi recovery and subsequent overshoot during reperfusion in myocytes pretreated with DOG (shown in Fig. 6B) were completely abolished by pretreatment with the PKC inhibitor chelerythrine.

Effects of BDM on the reduced diastolic cell length. At the end of each experiment in the protocols using SNARF 1 (n = 62), myocytes were superfused with Ca²⁺-free HEPES-buffered solution containing 30 mM BDM, 2 mM EGTA, and 1 µM thapsigargin for the purpose of pH_i calibration. A representative tracing illustrating the effect of BDM, EGTA, and thapsigargin on the reversal of the reperfusion-induced contracture in SNARF 1-loaded myocytes not treated with DOG is shown in Fig. 10. In the myocytes that showed reperfusion-induced contracture of <10% of the baseline diastolic cell length (7/62 myocytes, 3.7 ± 0.9% of baseline diastolic cell length), the reduced diastolic cell length was restored toward baseline by 51.5 ± 8.3% with the superfusion of the Ca²⁺-free HEPES-buffered solution containing BDM, EGTA, and thapsigargin. On the other hand, in the myocytes that showed reperfusion-induced contracture of >10% of the baseline diastolic cell length (55/62 myocytes, 47.1 ± 3.2% of baseline diastolic cell length), the reduced diastolic cell length was restored by only 5.4 ± 1.4% with the superfusion of the Ca²⁺-free HEPES-buffered solution containing BDM, EGTA, and thapsigargin. These results suggest that the reduction in diastolic cell length during reperfusion was largely irreversible.

View larger version (10K): [in this window] [in a new window]   Fig. 10.   A representative tracing showing effects of 2,3-butanedione monoxime (BDM), EGTA, and thapsigargin on reperfusion-induced contracture in an untreated myocyte loaded with SNARF 1. Reduced diastolic cell length during reperfusion was restored only 5% toward baseline with superfusion of Ca2+-free HEPES-buffered solution containing 30 mM BDM, 2 mM EGTA, and 1 µM thapsigargin.

Effects of BDM administered before and during I/R. To examine whether the reperfusion-induced contracture could be seen in the presence of BDM, additional experiments (n = 8 for each) were performed in which 10 mM BDM was initiated 5 min before the onset of simulated ischemia and continued throughout I/R. The reperfusion-induced contracture was almost abolished in both the DOG-pretreated and untreated myocytes (8.1 ± 1.8 vs. 6.3 ± 1.4% of baseline diastolic cell length, respectively, not significant). These data suggest that the reperfusion-induced contracture can be prevented when the contractile activation during I/R is inhibited.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION REFERENCES

We demonstrated that pretreatment with DOG attenuated the degree of contracture during reperfusion in isolated, paced rat ventricular myocytes. We also demonstrated that the pH_i overshoot during reperfusion was significantly delayed in the DOG-pretreated myocytes compared with the untreated myocytes and that there is a significant positive correlation between the degree of contracture and the time to pH_i 7.30 from the onset of reperfusion. Furthermore, the effects of DOG pretreatment on both the cell contracture and the pH_i overshoot were completely abolished by a highly selective PKC inhibitor, chelerythrine. These data suggest that the attenuation of I/R injury by DOG pretreatment is PKC dependent and raise the possibility that this cardioprotective effect may be related to the delayed pH_i overshoot during reperfusion, although this significant correlation does not necessarily indicate a cause-and-effect relationship. This is the first report showing the effects of PKC preactivation on cell motion and ionic alterations during I/R in isolated myocytes that were electrically stimulated throughout the course of the experiment.

The results in the present study are in agreement with several earlier reports. Kitakaze et al. reported that temporary acidosis during early reperfusion prevented myocardial stunning in isolated, perfused ferret hearts (24) and limited the myocardial infarct size in open-chest dogs (23). Nishida et al. (33) showed that temporary acidosis during early reperfusion prevented the development of hypercontracture at reperfusion in isolated rat myocytes. In those reports, the attenuation of Ca²⁺ overload via Na⁺/H⁺ exchanger and Na⁺/Ca²⁺ exchanger was proposed as one possible mechanism of the protective effect of temporary acidosis during reperfusion. However, in the present study there were no differences in either end-diastolic or peak-systolic [Ca²⁺]_i during reperfusion between the myocytes with and without DOG pretreatment. The attenuation of Ca²⁺ overload, therefore, cannot explain the beneficial effect of DOG pretreatment on the development of hypercontracture during I/R in our present study. In the earlier studies, myocytes were temporarily subjected to extracellular acidosis during the early phase of reperfusion ("acidic reperfusion"). On the other hand, we used a buffer with a pH of 7.40 to reperfuse both myocytes pretreated with DOG and those without pretreatment. The pH of the extracellular fluid, which is one of the major determinant factors that regulate Na⁺/H⁺ exchanger and other pH_i regulatory mechanisms, was identical in both groups in the present study. This might partially explain the different results regarding the Ca²⁺ load between the earlier studies and ours.

BDM is known to uncouple the contractile activity from Ca²⁺ transients (15, 38). In the present study, at the end of each pH_i experiment, myocytes were exposed to a Ca²⁺-free solution containing 30 mM BDM, 2 mM EGTA, and 1 µM thapsigargin for the purpose of pH_i calibration. The Ca²⁺-free solution containing 30 mM BDM restored only 5.4 ± 1.4% of the reperfusion-induced contracture in the myocytes that showed reperfusion-induced contracture of >10% of the baseline diastolic cell length (Fig. 10). These data indicate that BDM does not restore the reduced cell length once cell contracture has developed after reperfusion and suggest that the reduction in cell length after reperfusion shown in the present study was irreversible. In contrast, the reperfusion-induced contracture was almost abolished in both the DOG-pretreated and untreated myocytes when the superfusion with BDM was initiated before the onset of ischemia and continued throughout I/R. These results support the idea that the reperfusion-induced contracture may be induced secondarily to the strong contractile activation. This strong contractile activation may be explained by an increase in the myofilament responsiveness to Ca²⁺ because the increase in pH_i elicits an increase in the myofilament responsiveness to Ca²⁺ (2, 12, 45). It is possible that irreversible structural distortions, which may occur when the contractile force exceeds the reversible deformability of such components of the cell as the cytoskeleton, were smaller in the DOG-pretreated myocytes than in the control myocytes because of the delayed pH_i overshoot in the former, even though there was no difference in [Ca²⁺]_i between the two groups.

The involvement of rigor bonds is also one of the possible explanations for the myocyte contracture. If the rigor bonds played a major role in the irreversible contracture, however, the reduction in diastolic cell length would become overt during simulated ischemia, when ATP production should be lowest. It is unlikely, therefore, that the rigor bonds play a predominant role in the irreversible contracture. Osmotic swelling has also been raised as a possible explanation for the myocyte contracture. Ruiz-Meana et al. (39) reported that osmotic swelling disrupted the sarcolemma of viable myocytes during reperfusion using metabolic inhibition and hypotonic solution. In our present study, however, because BDM almost abolished the cell contracture during I/R when it was initiated before the onset of ischemia and continued throughout I/R, osmotic swelling by itself may not be able to explain the cell contracture.

The mechanism of the delay in pH_i overshoot induced by DOG pretreatment is still unclear. The delay in pH_i overshoot was completely abolished by the specific PKC inhibitor chelerythrine, which suggests that the process is PKC dependent. The pH_i increased significantly after the superfusion with chelerythrine (Fig. 9A), which suggests the existence of a PKC-dependent pH_i regulatory mechanism. This unknown mechanism may play an important role in the delay in pH_i overshoot during reperfusion in the DOG-pretreated myocytes. On the other hand, PKC is reported to phosphorylate both Na⁺/H⁺ exchanger and vacuolar proton ATPase, which elicit intracellular alkalinization (32, 34, 46). Gottlieb et al. reported that a target of PKC in mediating the cardioprotective effect of ischemic preconditioning is the activation of vacuolar proton ATPase with a resultant attenuation of the intracellular acidification during simulated ischemia and the subsequent prevention of apoptotic cell death in isolated quiescent rabbit (13) and rat (22) myocytes. If Na⁺/H⁺ exchanger or vacuolar proton ATPase is activated by PKC, however, the pH_i recovery during reperfusion would become faster rather than delayed, as in our results. Moreover, there was no difference in the pH_i during ischemia between the DOG-pretreated and untreated myocytes in the present study. Therefore, Na⁺/H⁺ exchanger or vacuolar proton ATPase may not play a predominant role in our model. Depolarization of the sarcolemmal membrane and a resultant twitch contraction greatly influence the homeostasis of intracellular ions and the cell integrity. Differences between the quiescent and paced myocytes during the protocol may explain the different results.

In the present study, after the onset of reperfusion, the pH_i showed a transient overshoot beyond the pH_i during the baseline superfusion. Silverman et al. (5, 41) also reported a transient rebound alkalosis at reoxygenation in studies using isolated cardiac myocytes and SNARF 1. On the other hand, in past studies using ³¹P NMR spectroscopy in isolated hearts, a transient overshoot in pH_i after the onset of reperfusion was not observed (27, 44). This may be because the NMR method used in those studies might not have detected the rapid changes in pH_i immediately after the onset of reperfusion due to its limited time resolution. Another possibility is that the myocytes exposed to rapid and direct environmental changes show an excessive response because our isolated myocyte model had no interstitial components.

Limitations. There are some possible limitations in our present study. First, there was some difference in the degree of reperfusion-induced contracture between the myocytes loaded with indo 1 and those loaded with SNARF 1, although DOG pretreatment significantly attenuated the degree of contracture in both fluorescence indicator groups. Although we used indo 1 at a concentration as low as possible that could still provide enough fluorescence signal for analysis, we could not exclude the possibility of the Ca²⁺-buffering effect of indo 1 (14). Second, the calibration curve of SNARF 1 was not linear at extremely low pH. It was, however, nearly linear at the pH range from 6.4 to 7.8 used in this study (data not shown). Because, as shown in Fig. 6A, the minimum pH_i during simulated ischemia tended to be lower in the DOG-pretreated myocytes than in the control myocytes, the possibility that SNARF 1 cannot detect the attenuation of acidosis during simulated ischemia in the DOG-pretreated myocytes seems unlikely. Third, Light et al. (26) and Liu et al. (28) reported that PKC activation resulted in increased ATP-sensitive K⁺ channel (K_ATP) current in isolated rabbit ventricular myocytes. Because there were no differences in cytosolic [Ca²⁺]_i levels between the myocytes with and those without the DOG pretreatment in our present study, it is unlikely that K_ATP plays a predominant role in the protective effect by DOG. Nevertheless, the possibility that K_ATP not in the sarcolemmal membrane but in the mitochondrial membrane was involved cannot be excluded (29, 40). Fourth, the twitch amplitude during reperfusion was similar to the baseline value in each group, although the amplitude of the Ca²⁺ transient during reperfusion was smaller than that during the baseline measurement. It is possible that the myofilament responsiveness to Ca²⁺ was increased after reperfusion compared with that in the baseline condition. However, because we could not control the sarcomere length, which should be changed greatly by the structural distortions during I/R, the changes in myofilament responsiveness to Ca²⁺ during reperfusion could not be evaluated in the present study (1). Finally, there are many parameters that fluctuate during I/R. However, we focused only on the cell motion, [Ca²⁺]_i, and pH_i. Furthermore, the significant correlation between the degree of contracture and the time to pH_i 7.30 during reperfusion does not necessarily indicate a cause-and-effect relationship. The possibility that the difference in pH_i change between the DOG-pretreated and untreated myocytes during reperfusion is an epiphenomenon cannot be totally excluded.